Skip to main content

david ansell

Followers

0

Following

1

Public Views

Interests

Uploads

Papers by david ansell

A community effort to reduce the black/white breast cancer mortality disparity in Chicago

Cancer Causes & Control, 2009

Background The Metropolitan Chicago Breast Cancer Taskforce was formed to address a growing black... more Background The Metropolitan Chicago Breast Cancer Taskforce was formed to address a growing black/white breast cancer mortality disparity in Chicago. The Taskforce explored three hypotheses: black women in Chicago receive fewer mammograms, black women receive mammograms of inferior quality, and black women have inadequate access to quality of treatment for breast cancer. Methods A total of 102 individuals from 74 Chicago area organizations participated in the Task Force participating in three work groups from January to September 2007. The work groups held focus groups of providers, organized town hall meetings in four Chicago communities, gathered black/white breast cancer mortality data for Chicago, the United States, and New York City, and conducted a mammography capacity and quality survey of mammography facilities. Results Chicago's black and white breast cancer mortality rates were the same in 1980. By the late 1990s, a substantial disparity was present, and by 2005, the black breast cancer mortality rate was 116% higher than the white rate. In 2007, 206,000 screening mammograms were performed for women living in Chicago, far short of the 588,000 women in the 40-69 age range in Chicago. Facilities that served predominately minority women were less likely to be academic or private institutions (p \ .03), less likely to have digital mammography (p \ .003), and less likely to have dedicated breast imaging specialists reading the films (p \ .003). Black women and providers serving them reported significant difficulties in accessing needed care for breast cancer screening and treatment. Conclusion There are significant access barriers to high quality mammography and treatment services that could be contributing to the mortality differences in Chicago. A metropolitan wide taskforce has been established to address the disparity.

Diversity and Disparity: GIS and Small-Area Analysis in Six Chicago Neighborhoods

Journal of Medical Systems, 2004

Small-area analysis in health is essential in uncovering local-level disparities often masked by ... more Small-area analysis in health is essential in uncovering local-level disparities often masked by health estimates for large areas (e.g., cities, counties, states). In this context, 14 health status indicators (HSIs) were examined for six Chicago community areas that reflect the substantial diversity of the city. HSIs were compared over time (from 1989-90 to 1999-2000) and across community areas. Important disparities among these community areas in mortality rates, birth outcomes, and infectious diseases were found. In many cases the disparities were in the expected direction with the richest and predominantly White community area experiencing the lowest rates. However, some surprises did manifest themselves. For example, only the poorest community area experienced a statistically significant decline in the infant mortality rate. Since so much of attention is now being paid to reducing and eliminating these disparities, it is important to examine their existence to better understand how to minimize them.

Rethinking the Mission of Health Systems: Improving Community Health as Anchor Institutions

Oxford Medicine Online

This chapter describes a new and compelling role for hospitals and health systems as anchor insti... more This chapter describes a new and compelling role for hospitals and health systems as anchor institutions. Representing significant economic and human capital resources, anchor institutions are grounded, sound business practices, and in the realities of the local communities they serve, they are well placed to effectively align, leverage, and invest resources to improve the health of their communities. An important and highly effective partner for health, examples of successful institutions that have adopted this approach are described, as is a national network to help hospitals and other institutions adopt and advance an anchor mission within their organizations and communities.

As Natural as the Air Around Us": On the Origin and Development of the Concept of Structural Violence in Health Research

International journal of health services : planning, administration, evaluation, Oct 1, 2018

This article examines the concept of "structural violence." Originating in the work of ... more This article examines the concept of "structural violence." Originating in the work of Johan Galtung in 1969 and popularized by Paul Farmer, structural violence is increasingly invoked in health literature. It is a complex concept - rich in its explanatory potential but vague in its operational definition and arguably limited in its theoretical precision. Its potential lies in the focus it gives to the deep structural roots of health inequities; in contrast to the more passive term "social determinants of health," structural violence explicitly identifies social, economic, and political systems as the causes of the causes of poor health. It is also evocative in its framing of health inequities as an act of violence. Yet the formulation of structure used in this literature is largely atheoretical and, by extension, apolitical. Development of the concept hinges on clarifying the precise aspects of structure it points to (perhaps through using the concept in conjunc...

Association of community-level inequities and premature mortality: Chicago, 2011–2015

Journal of Epidemiology and Community Health

BackgroundSubstantial disparities in life expectancy exist between Chicago’s 77 defined community... more BackgroundSubstantial disparities in life expectancy exist between Chicago’s 77 defined community areas, ranging from approximately 69 to 85 years. Prior work in New York City and Boston has shown that community-level racial and economic segregation as measured by the Index of Concentration at the Extremes (ICE) is strongly related to premature mortality. This novel metric allows for the joint assessment of area-based income and racial polarisation. This study aimed to assess the relationships between racial and economic segregation and economic hardship with premature mortality in Chicago.MethodsAnnual age-adjusted premature mortality rates (deaths <65 years) from 2011 to 2015 were calculated for Chicago’s 77 community areas. ICE measures for household income (<US$25 000 vs ≥US$100 000), race (black vs non-Hispanic white), combined ICE measure incorporating income and race, and hardship index were calculated from 2015 American Community Survey 5-year estimates.ResultsAverage ...

Racial/ethnic minority segregation and low birth weight in five North American cities

Ethnicity & health, Jan 27, 2018

Comparisons of communities across cities are rare in social epidemiology. Our prior work explorin... more Comparisons of communities across cities are rare in social epidemiology. Our prior work exploring racial/ethnic segregation and the prevalence of low birth weight (LBW) in communities from two large urban cities showed a strong relationship in Chicago and a very weak relationship in Toronto. This study extends that work by examining the association between racial/ethnic minority segregation and LBW in total of 307 communities in five North American cities: Baltimore, Boston, Chicago, Philadelphia, and Toronto. We used Pearson correlation coefficients and OLS regression models to examine potential variability in the association between racial/ethnic minority segregation and LBW, controlling for community-level unemployment. In a combined model with community-level data from all cities, a 10% increase in minority composition is associated with a 0.7% increase in LBW. While racial/ethnic minority segregation and unemployment are not associated with LBW in Toronto, these social determi...

Disparity in breast cancer mortality by age and geography in 10 racially diverse US cities

Cancer epidemiology, Apr 1, 2018

Assess geographic variation in breast cancer racial mortality disparity by age cohorts in US and ... more Assess geographic variation in breast cancer racial mortality disparity by age cohorts in US and ten cities with large African American populations. Non-Hispanic Black (NHB) and Non-Hispanic White (NHW) female breast cancer mortality rates and NHB:NHW rate ratio (RR) (disparity) were calculated by four age group categories: <40, 40-49, 50-64 and 65+ with time period 1999-2013. In all 10 cities and the US, the most pronounced breast cancer disparities, measured by RR, were seen among younger women. In age group <40, the RR ranges from 1.71 in Houston to 5.37 in Washington, DC. For age group 50-64, the disparity was less pronounced, ranging from 1.24 in New York to 1.72 in Chicago. For 65+ age group, there was wide city to city variation in breast cancer mortality disparity. Three cities had higher mortality for NHW compared to NHB; Baltimore 0.78, Washington DC 0.94 and New York 0.98. One city had no statistically significant racial variation in breast cancer mortality in this ...

Reply to “Relative Utility or Marginal Positive Predictive Values Accounting for Overdiagnosis Should Guide Optimal Recall Rates”

American Journal of Roentgenology

The goal of mammography screening is to reduce breast cancer mortality. We appreciate that there ... more The goal of mammography screening is to reduce breast cancer mortality. We appreciate that there is controversy about which types of ductal carcinoma in situ (DCIS) or invasive breast cancers will result in mortality. Until this issue is resolved, however, the job of the breast imager remains to detect breast cancers when they are small. The finding in our academic practice is that recall rates of 12% to less than 14% provide an optimal balance between the detection of significant cancer and unnecessary testing [1]. On the basis of Breast Cancer Surveillance Consortium (BCSC) data, Schell et al. [2] recommend an acceptable recall rate of 10%. On the basis of our study data [1], a 10% recall rate is too low and significant cancers in our center would have been missed. There are three possible reasons why our findings [1] differed from those of the BCSC study [2]. First, our study was performed from 2007 to 2012, in the digital mammography era, whereas the BCSC study was conducted from 1996 to 2001. It is likely that the use of more advanced technology contributed to both increased cancer detection and increased recall rates. Second, the BCSC data analysis used a different method than our study. Schell et al. [2] advocated a threshold of 100 addi

Reply to “Finding the Best Recall and Cancer Detection Rates for Screening Mammography”

American Journal of Roentgenology

Process improvement in breast cancer care: Is mammography volume associated with a greater need for process improvement?

Journal of Clinical Oncology, 2013

45 Background: Chicago black women are 62% more likely to die from breast cancer than white women... more 45 Background: Chicago black women are 62% more likely to die from breast cancer than white women. Previous data from 39 Chicago hospitals suggested significant variation in mammography quality (Chicago Breast Cancer Quality Consortium, 2010). We developed process improvement recommendations for sites that participated in our care process assessment (Weldon CB, et al, ASCO--6120-2012). This study was funded through a generous grant from the Susan G. Komen for the Cure Foundation. We compared improvement needs between high and low patient volume institutions. Methods: Using Deming’s PDCA cycle for continuous improvement, we created care process improvement recommendations for 25 Chicago institutions with the patient base averaging more than 50% minority patients (20 community, 3 academic and 2 public hospitals). Low mammography volume (< 5,000 mammograms/ year) was reported by 12 of the 25 sites. Recommendations are based on analysis across sites, literature, and input from instit...

Recall and Cancer Detection Rates for Screening Mammography: Finding the Sweet Spot

American Journal of Roentgenology, 2016

The purpose of this study is to identify the optimal screening mammography recall rate range on t... more The purpose of this study is to identify the optimal screening mammography recall rate range on the basis of cancer detection rates among breast imaging specialists at an academic institution. Medical outcome audit data collected in accordance with the Mammography Quality Standards Act from September 1, 2007, through August 31, 2012, were reviewed. Cancer detection rates were calculated from 984 screen-detected cancers identified in 188,959 total digital screening mammograms. The percentages of minimally invasive and early-stage cancers were also calculated. The 75 annual recall rates were analyzed two ways. First, they were separated into recall groups to assess cancer detection rate variation by the recall categories using rate ratios: less than 10%, 10% to less than 12%, 12% to less than 14%, and 14% or higher. Next, a linear regression with bootstrap bias correction was performed to assess changes in cancer detection rate with each unit increase in the recall rate up to 20%, with the recall category of less than 7% taken as reference. Annual cancer detection rates for a physician were grouped according to annual percentage recall rate. Statistically significantly higher cancer detection rates were seen for recall rates 12% or higher, with rate ratios of 1.75 (95% CI, 1.40-2.19) and 2.06 (95% CI, 1.72-2.46) for the recall groups 12% to less than 14% and 14% and higher, respectively, compared with the less than 10% group. When taking the category 12% to less than 14% as the reference, there were no statistically significant differences between recall groups 12% to less than 14% and 14% or higher in cancer detection rate. A statistically significant increase in the cancer detection rate with each unit increase in the recall rate was seen only for recall rates 12% or higher. These observations suggest that the sweet spot for optimal cancer detection is in the recall rate range 12% to less than 14% with the incremental benefit above this to be relatively small. A recall rate less than 10% may be too low.

The CDC rules that associate urinary catheter use with urinary tract infection fail to measure what we want

ABSTRACT Background: The Center for Disease Control (CDC) defines a catheter-associated urinary t... more ABSTRACT Background: The Center for Disease Control (CDC) defines a catheter-associated urinary tract infection (CA-UTI) as one occurring within 7 days of the presence of a urinary catheter (UC) rather than later than 7 days. We examined the odds of a potential UTI based on this time cut-off. Methods: We merged data on patients hospitalized between July, 2007 and April, 2008 who had both a UC and urine cultures. We excluded patients with ureteral stents, suprapubic catheters, nephrostomy tubes or ileoconduits. We then classified patients based on urine culture results: greater than 105 true urine pathogen (CDC definition of catheter-associated asymptomatic bacteruria); contaminant; or, negative culture. We compared the odds of isolating a true urine pathogen within 7 days or greater than 7 days of a UC. Results: We reviewed 3791 culture results from 1827 patients. The odds for having asymptomatic bacteruria were the same regardless of the timing of urine culture (within/not within 7 days of a UC) [asymptomatic bacteruria OR 0.93 (95% CI 0.73-1.20)]. Conclusion: Our data does not support the CDC CA-UTI criterion for timing of a urine culture. We did not detect a difference in the risk of asymptomatic bacteruria between cultures drawn within or not within 7 days of a UC. This data may suggest that removal of a UC does not reduce the risk of UTI

Accuracy of electronic data in estimating the number of inpatients with a urinary catheter

ABSTRACT Background: The Center for Disease Control considers hospital rates of catheter-associat... more ABSTRACT Background: The Center for Disease Control considers hospital rates of catheter-associated urinary tract infections (CA-UTI) an important clinical measurement. The electronic medical record (EMR) purportedly represents a robust database from which one may efficiently calculate rates of CA-UTI. We sought to determine the utility of specific electronic data that might identify patients at risk for CA-UTI, namely those with a urinary catheter (UC). Methods: We identified all patients admitted to Rush University Medical Center in Chicago, IL who had nursing flowsheet data for UC care from July 1, 2008 through March 31, 2009. We categorized patients by the method of catheter identification: charge for catheter; electronic order for catheter placement or removal; or, electronic nursing flowsheet entry. We then determined the proportion of patients identified with UC using each method compared to a gold standard. We considered nursing flowsheet data as the gold standard after a pilot study revealed 100 percent agreement between this data and contemporaneous bedside observation of a UC. Results: We reviewed 2113 unique inpatient encounters as identified by flowsheet data. Neither UC charges nor UC orders provided good estimates of the presence of a UC: 130 charges for UC (6.15% of all flowsheet encounters); 863 UC orders (40.84% of flowsheet encounters. Conclusion: Urinary catheter orders and charge data fail to identify the majority of inpatients with a urinary catheter. Underestimation of at-risk patients will lead to overestimation of CA-UTI rates. Nursing flowsheet data provides the most accurate estimate of patients at risk for CA-UTI.

2005. The effect of cocaine and heroin use on intubation rates and hospital utilization in patients with acute asthma exacerbations

Chest

CERN: A city-wide cross-hospital collaborative for infectious diseases preparedness

Clinical Infectious Diseases, 2015

The 2014-2015 Ebola virus disease (EVD) epidemic and international public health emergency has be... more The 2014-2015 Ebola virus disease (EVD) epidemic and international public health emergency has been referred to as a "black swan" event, or an event that is unlikely, hard to predict, and highly impactful once it occurs. The Chicago Ebola Response Network (CERN) was formed in response to EVD and is capable of receiving and managing new cases of EVD, while also laying the foundation for a public health network that can anticipate, manage, and prevent the next black swan public health event. By sharing expertise, risk, and resources among 4 major academic centers, Chicago created a sustainable network to respond to the latest in a series of public health emergencies. In this respect, CERN is a roadmap for how a region can prepare to respond to public health emergencies, thereby preventing negative impacts through planning and implementation.

Abstract A84: Do patient follow-up improvements, at hospitals caring for medically underserved patients, impact no-show rates

Cancer Epidemiology Biomarkers & Prevention, 2014

Background: In Chicago Black women are 62% more likely to die from breast cancer than White women... more Background: In Chicago Black women are 62% more likely to die from breast cancer than White women. Previous data from 39 Chicago hospitals suggested the existence of significant variation in mammography quality across facilities (Chicago Breast Cancer Quality Consortium, 2010). We developed process improvement recommendations for sites that participated in our care process assessment (Weldon CB, et al, ASCO-Abstract-6120-2012). This study was funded through a generous grant from the Susan G. Komen for the Cure Foundation. Appointment reminder calls and follow-up calls to patients that do not attend their appointments are effective in increasing breast cancer screening and diagnosis completion rates of (Goel A, et al, JHCPU 2008). We provided technical assistance to implement breast imaging care process improvements and compared pre and post process improvement appointment no-show rates and follow-up call rates at 14 sites. Procedures: Using Deming9s PDCA cycle for continuous improvement, we created care process improvement recommendations for 27 Chicago institutions with the patient base averaging more than 50% medically underserved patients (22 community, 3 academic and 2 public hospitals), 17 of the sites are safety net institutions. Recommendations were based on analysis across sites, literature, and input from institution staff. We provided 14 of the sites with technical assistance to implement “rapid cycle” care process improvements (RCI sites). Sites were selected for technical assistance based on their need and their commitment to improving breast imaging care processes. Thematic and statistical analyses were performed using simple frequencies, and McNemar9s test. Summary of Results: Detailed results are shown in the table below. Technical assistance was utilized by 14 of 17 safety net sites (RCI sites) to which it was offered. 10 sites adjusted their processes to conduct appointment reminder calls to patients. 12 sites implemented processes to conduct follow up phone calls with patients who missed an appointment, with a focus on answering patient questions and to schedule a new appointment. 6 sites adjusted their processes to conduct phone calls to inform patients of abnormal breast screening results (BIRADS 0, 4, 5) in addition to sending each patient a letter (MQSA requires at least a letter). Working with the 14 sites resulted in an improvement in no-show rates from 29% to 21%. Appointment reminder call: baseline 48% (13/27) all sites, 21% (3/14) RCI sites post-improvement 85% (23/27) all sites, 93% (13/14) RCI sites pvalue = 0.0044, Chi squared: 8.100 w 1 degree of freedom Call no-show patients : baseline 19% (5/27) all sites, 0% (0/14) RCI sites post-improvement 63% (17/27) all sites, 86% (12/14) RCI sites pvalue = 0.0015, Chi squared: 10.083 w 1 degree of freedom Call to follow up with patients who have abnormal results: baseline 56% (15/27) all sites, 50% (7/14) RCI sites post-improvement 78% (21/27) all sites, 93% (13/14) RCI sites pvalue = 0.0412, Chi squared: 4.167 w 1 degree of freedom No show rate for screening mammograms baseline 22% all sites, 29% RCI sites : post-improvement 18% all sites, 21% RCI sites Conclusions: Conducting phone calls to remind patients about appointments improves no-show rates at sites that care for the medically underserved. Further analysis may show a reduction in loss-to-follow-up for sites that implement follow up phone calls to patients who miss appointments and to patients who need additional diagnostic assessment. Findings from this study have generalizable application to health facilities beyond breast imaging sites. Given persistent and growing disparities in health outcomes for vulnerable populations and the limited resource availability; these findings suggest that implementation of basic patient tracking strategies have substantial benefit to improve patient outcomes and health care quality. Citation Format: Christine B. Weldon, Teena L. Francois, Julia R. Trosman, Betty Roggenkamp, Danielle M. Dupuy, Jimmie T. Knight, David A. Ansell, Anne Marie Murphy. Do patient follow-up improvements, at hospitals caring for medically underserved patients, impact no-show rates. [abstract]. In: Proceedings of the Sixth AACR Conference: The Science of Cancer Health Disparities; Dec 6–9, 2013; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2014;23(11 Suppl):Abstract nr A84. doi:10.1158/1538-7755.DISP13-A84

Illinois law opens door to kidney transplants for undocumented immigrants

Health affairs (Project Hope), 2015

Bias, black lives, and academic medicine

The New England journal of medicine, Jan 19, 2015

Federal anti-patient-dumping provisions: the first decade

Annals of emergency medicine, 1996

Colorectal cancer screening: trends in an urban primary care setting

American Journal of Gastroenterology, 2002

A community effort to reduce the black/white breast cancer mortality disparity in Chicago

Cancer Causes & Control, 2009

Background The Metropolitan Chicago Breast Cancer Taskforce was formed to address a growing black... more Background The Metropolitan Chicago Breast Cancer Taskforce was formed to address a growing black/white breast cancer mortality disparity in Chicago. The Taskforce explored three hypotheses: black women in Chicago receive fewer mammograms, black women receive mammograms of inferior quality, and black women have inadequate access to quality of treatment for breast cancer. Methods A total of 102 individuals from 74 Chicago area organizations participated in the Task Force participating in three work groups from January to September 2007. The work groups held focus groups of providers, organized town hall meetings in four Chicago communities, gathered black/white breast cancer mortality data for Chicago, the United States, and New York City, and conducted a mammography capacity and quality survey of mammography facilities. Results Chicago's black and white breast cancer mortality rates were the same in 1980. By the late 1990s, a substantial disparity was present, and by 2005, the black breast cancer mortality rate was 116% higher than the white rate. In 2007, 206,000 screening mammograms were performed for women living in Chicago, far short of the 588,000 women in the 40-69 age range in Chicago. Facilities that served predominately minority women were less likely to be academic or private institutions (p \ .03), less likely to have digital mammography (p \ .003), and less likely to have dedicated breast imaging specialists reading the films (p \ .003). Black women and providers serving them reported significant difficulties in accessing needed care for breast cancer screening and treatment. Conclusion There are significant access barriers to high quality mammography and treatment services that could be contributing to the mortality differences in Chicago. A metropolitan wide taskforce has been established to address the disparity.

Diversity and Disparity: GIS and Small-Area Analysis in Six Chicago Neighborhoods

Journal of Medical Systems, 2004

Small-area analysis in health is essential in uncovering local-level disparities often masked by ... more Small-area analysis in health is essential in uncovering local-level disparities often masked by health estimates for large areas (e.g., cities, counties, states). In this context, 14 health status indicators (HSIs) were examined for six Chicago community areas that reflect the substantial diversity of the city. HSIs were compared over time (from 1989-90 to 1999-2000) and across community areas. Important disparities among these community areas in mortality rates, birth outcomes, and infectious diseases were found. In many cases the disparities were in the expected direction with the richest and predominantly White community area experiencing the lowest rates. However, some surprises did manifest themselves. For example, only the poorest community area experienced a statistically significant decline in the infant mortality rate. Since so much of attention is now being paid to reducing and eliminating these disparities, it is important to examine their existence to better understand how to minimize them.

Rethinking the Mission of Health Systems: Improving Community Health as Anchor Institutions

Oxford Medicine Online

This chapter describes a new and compelling role for hospitals and health systems as anchor insti... more This chapter describes a new and compelling role for hospitals and health systems as anchor institutions. Representing significant economic and human capital resources, anchor institutions are grounded, sound business practices, and in the realities of the local communities they serve, they are well placed to effectively align, leverage, and invest resources to improve the health of their communities. An important and highly effective partner for health, examples of successful institutions that have adopted this approach are described, as is a national network to help hospitals and other institutions adopt and advance an anchor mission within their organizations and communities.

As Natural as the Air Around Us": On the Origin and Development of the Concept of Structural Violence in Health Research

International journal of health services : planning, administration, evaluation, Oct 1, 2018

This article examines the concept of "structural violence." Originating in the work of ... more This article examines the concept of "structural violence." Originating in the work of Johan Galtung in 1969 and popularized by Paul Farmer, structural violence is increasingly invoked in health literature. It is a complex concept - rich in its explanatory potential but vague in its operational definition and arguably limited in its theoretical precision. Its potential lies in the focus it gives to the deep structural roots of health inequities; in contrast to the more passive term "social determinants of health," structural violence explicitly identifies social, economic, and political systems as the causes of the causes of poor health. It is also evocative in its framing of health inequities as an act of violence. Yet the formulation of structure used in this literature is largely atheoretical and, by extension, apolitical. Development of the concept hinges on clarifying the precise aspects of structure it points to (perhaps through using the concept in conjunc...

Association of community-level inequities and premature mortality: Chicago, 2011–2015

Journal of Epidemiology and Community Health

BackgroundSubstantial disparities in life expectancy exist between Chicago’s 77 defined community... more BackgroundSubstantial disparities in life expectancy exist between Chicago’s 77 defined community areas, ranging from approximately 69 to 85 years. Prior work in New York City and Boston has shown that community-level racial and economic segregation as measured by the Index of Concentration at the Extremes (ICE) is strongly related to premature mortality. This novel metric allows for the joint assessment of area-based income and racial polarisation. This study aimed to assess the relationships between racial and economic segregation and economic hardship with premature mortality in Chicago.MethodsAnnual age-adjusted premature mortality rates (deaths <65 years) from 2011 to 2015 were calculated for Chicago’s 77 community areas. ICE measures for household income (<US$25 000 vs ≥US$100 000), race (black vs non-Hispanic white), combined ICE measure incorporating income and race, and hardship index were calculated from 2015 American Community Survey 5-year estimates.ResultsAverage ...

Racial/ethnic minority segregation and low birth weight in five North American cities

Ethnicity & health, Jan 27, 2018

Comparisons of communities across cities are rare in social epidemiology. Our prior work explorin... more Comparisons of communities across cities are rare in social epidemiology. Our prior work exploring racial/ethnic segregation and the prevalence of low birth weight (LBW) in communities from two large urban cities showed a strong relationship in Chicago and a very weak relationship in Toronto. This study extends that work by examining the association between racial/ethnic minority segregation and LBW in total of 307 communities in five North American cities: Baltimore, Boston, Chicago, Philadelphia, and Toronto. We used Pearson correlation coefficients and OLS regression models to examine potential variability in the association between racial/ethnic minority segregation and LBW, controlling for community-level unemployment. In a combined model with community-level data from all cities, a 10% increase in minority composition is associated with a 0.7% increase in LBW. While racial/ethnic minority segregation and unemployment are not associated with LBW in Toronto, these social determi...

Disparity in breast cancer mortality by age and geography in 10 racially diverse US cities

Cancer epidemiology, Apr 1, 2018

Assess geographic variation in breast cancer racial mortality disparity by age cohorts in US and ... more Assess geographic variation in breast cancer racial mortality disparity by age cohorts in US and ten cities with large African American populations. Non-Hispanic Black (NHB) and Non-Hispanic White (NHW) female breast cancer mortality rates and NHB:NHW rate ratio (RR) (disparity) were calculated by four age group categories: <40, 40-49, 50-64 and 65+ with time period 1999-2013. In all 10 cities and the US, the most pronounced breast cancer disparities, measured by RR, were seen among younger women. In age group <40, the RR ranges from 1.71 in Houston to 5.37 in Washington, DC. For age group 50-64, the disparity was less pronounced, ranging from 1.24 in New York to 1.72 in Chicago. For 65+ age group, there was wide city to city variation in breast cancer mortality disparity. Three cities had higher mortality for NHW compared to NHB; Baltimore 0.78, Washington DC 0.94 and New York 0.98. One city had no statistically significant racial variation in breast cancer mortality in this ...

Reply to “Relative Utility or Marginal Positive Predictive Values Accounting for Overdiagnosis Should Guide Optimal Recall Rates”

American Journal of Roentgenology

The goal of mammography screening is to reduce breast cancer mortality. We appreciate that there ... more The goal of mammography screening is to reduce breast cancer mortality. We appreciate that there is controversy about which types of ductal carcinoma in situ (DCIS) or invasive breast cancers will result in mortality. Until this issue is resolved, however, the job of the breast imager remains to detect breast cancers when they are small. The finding in our academic practice is that recall rates of 12% to less than 14% provide an optimal balance between the detection of significant cancer and unnecessary testing [1]. On the basis of Breast Cancer Surveillance Consortium (BCSC) data, Schell et al. [2] recommend an acceptable recall rate of 10%. On the basis of our study data [1], a 10% recall rate is too low and significant cancers in our center would have been missed. There are three possible reasons why our findings [1] differed from those of the BCSC study [2]. First, our study was performed from 2007 to 2012, in the digital mammography era, whereas the BCSC study was conducted from 1996 to 2001. It is likely that the use of more advanced technology contributed to both increased cancer detection and increased recall rates. Second, the BCSC data analysis used a different method than our study. Schell et al. [2] advocated a threshold of 100 addi

Reply to “Finding the Best Recall and Cancer Detection Rates for Screening Mammography”

American Journal of Roentgenology

Process improvement in breast cancer care: Is mammography volume associated with a greater need for process improvement?

Journal of Clinical Oncology, 2013

45 Background: Chicago black women are 62% more likely to die from breast cancer than white women... more 45 Background: Chicago black women are 62% more likely to die from breast cancer than white women. Previous data from 39 Chicago hospitals suggested significant variation in mammography quality (Chicago Breast Cancer Quality Consortium, 2010). We developed process improvement recommendations for sites that participated in our care process assessment (Weldon CB, et al, ASCO--6120-2012). This study was funded through a generous grant from the Susan G. Komen for the Cure Foundation. We compared improvement needs between high and low patient volume institutions. Methods: Using Deming’s PDCA cycle for continuous improvement, we created care process improvement recommendations for 25 Chicago institutions with the patient base averaging more than 50% minority patients (20 community, 3 academic and 2 public hospitals). Low mammography volume (< 5,000 mammograms/ year) was reported by 12 of the 25 sites. Recommendations are based on analysis across sites, literature, and input from instit...

Recall and Cancer Detection Rates for Screening Mammography: Finding the Sweet Spot

American Journal of Roentgenology, 2016

The purpose of this study is to identify the optimal screening mammography recall rate range on t... more The purpose of this study is to identify the optimal screening mammography recall rate range on the basis of cancer detection rates among breast imaging specialists at an academic institution. Medical outcome audit data collected in accordance with the Mammography Quality Standards Act from September 1, 2007, through August 31, 2012, were reviewed. Cancer detection rates were calculated from 984 screen-detected cancers identified in 188,959 total digital screening mammograms. The percentages of minimally invasive and early-stage cancers were also calculated. The 75 annual recall rates were analyzed two ways. First, they were separated into recall groups to assess cancer detection rate variation by the recall categories using rate ratios: less than 10%, 10% to less than 12%, 12% to less than 14%, and 14% or higher. Next, a linear regression with bootstrap bias correction was performed to assess changes in cancer detection rate with each unit increase in the recall rate up to 20%, with the recall category of less than 7% taken as reference. Annual cancer detection rates for a physician were grouped according to annual percentage recall rate. Statistically significantly higher cancer detection rates were seen for recall rates 12% or higher, with rate ratios of 1.75 (95% CI, 1.40-2.19) and 2.06 (95% CI, 1.72-2.46) for the recall groups 12% to less than 14% and 14% and higher, respectively, compared with the less than 10% group. When taking the category 12% to less than 14% as the reference, there were no statistically significant differences between recall groups 12% to less than 14% and 14% or higher in cancer detection rate. A statistically significant increase in the cancer detection rate with each unit increase in the recall rate was seen only for recall rates 12% or higher. These observations suggest that the sweet spot for optimal cancer detection is in the recall rate range 12% to less than 14% with the incremental benefit above this to be relatively small. A recall rate less than 10% may be too low.

The CDC rules that associate urinary catheter use with urinary tract infection fail to measure what we want

ABSTRACT Background: The Center for Disease Control (CDC) defines a catheter-associated urinary t... more ABSTRACT Background: The Center for Disease Control (CDC) defines a catheter-associated urinary tract infection (CA-UTI) as one occurring within 7 days of the presence of a urinary catheter (UC) rather than later than 7 days. We examined the odds of a potential UTI based on this time cut-off. Methods: We merged data on patients hospitalized between July, 2007 and April, 2008 who had both a UC and urine cultures. We excluded patients with ureteral stents, suprapubic catheters, nephrostomy tubes or ileoconduits. We then classified patients based on urine culture results: greater than 105 true urine pathogen (CDC definition of catheter-associated asymptomatic bacteruria); contaminant; or, negative culture. We compared the odds of isolating a true urine pathogen within 7 days or greater than 7 days of a UC. Results: We reviewed 3791 culture results from 1827 patients. The odds for having asymptomatic bacteruria were the same regardless of the timing of urine culture (within/not within 7 days of a UC) [asymptomatic bacteruria OR 0.93 (95% CI 0.73-1.20)]. Conclusion: Our data does not support the CDC CA-UTI criterion for timing of a urine culture. We did not detect a difference in the risk of asymptomatic bacteruria between cultures drawn within or not within 7 days of a UC. This data may suggest that removal of a UC does not reduce the risk of UTI

Accuracy of electronic data in estimating the number of inpatients with a urinary catheter

ABSTRACT Background: The Center for Disease Control considers hospital rates of catheter-associat... more ABSTRACT Background: The Center for Disease Control considers hospital rates of catheter-associated urinary tract infections (CA-UTI) an important clinical measurement. The electronic medical record (EMR) purportedly represents a robust database from which one may efficiently calculate rates of CA-UTI. We sought to determine the utility of specific electronic data that might identify patients at risk for CA-UTI, namely those with a urinary catheter (UC). Methods: We identified all patients admitted to Rush University Medical Center in Chicago, IL who had nursing flowsheet data for UC care from July 1, 2008 through March 31, 2009. We categorized patients by the method of catheter identification: charge for catheter; electronic order for catheter placement or removal; or, electronic nursing flowsheet entry. We then determined the proportion of patients identified with UC using each method compared to a gold standard. We considered nursing flowsheet data as the gold standard after a pilot study revealed 100 percent agreement between this data and contemporaneous bedside observation of a UC. Results: We reviewed 2113 unique inpatient encounters as identified by flowsheet data. Neither UC charges nor UC orders provided good estimates of the presence of a UC: 130 charges for UC (6.15% of all flowsheet encounters); 863 UC orders (40.84% of flowsheet encounters. Conclusion: Urinary catheter orders and charge data fail to identify the majority of inpatients with a urinary catheter. Underestimation of at-risk patients will lead to overestimation of CA-UTI rates. Nursing flowsheet data provides the most accurate estimate of patients at risk for CA-UTI.

2005. The effect of cocaine and heroin use on intubation rates and hospital utilization in patients with acute asthma exacerbations

Chest

CERN: A city-wide cross-hospital collaborative for infectious diseases preparedness

Clinical Infectious Diseases, 2015

The 2014-2015 Ebola virus disease (EVD) epidemic and international public health emergency has be... more The 2014-2015 Ebola virus disease (EVD) epidemic and international public health emergency has been referred to as a "black swan" event, or an event that is unlikely, hard to predict, and highly impactful once it occurs. The Chicago Ebola Response Network (CERN) was formed in response to EVD and is capable of receiving and managing new cases of EVD, while also laying the foundation for a public health network that can anticipate, manage, and prevent the next black swan public health event. By sharing expertise, risk, and resources among 4 major academic centers, Chicago created a sustainable network to respond to the latest in a series of public health emergencies. In this respect, CERN is a roadmap for how a region can prepare to respond to public health emergencies, thereby preventing negative impacts through planning and implementation.

Abstract A84: Do patient follow-up improvements, at hospitals caring for medically underserved patients, impact no-show rates

Cancer Epidemiology Biomarkers & Prevention, 2014

Background: In Chicago Black women are 62% more likely to die from breast cancer than White women... more Background: In Chicago Black women are 62% more likely to die from breast cancer than White women. Previous data from 39 Chicago hospitals suggested the existence of significant variation in mammography quality across facilities (Chicago Breast Cancer Quality Consortium, 2010). We developed process improvement recommendations for sites that participated in our care process assessment (Weldon CB, et al, ASCO-Abstract-6120-2012). This study was funded through a generous grant from the Susan G. Komen for the Cure Foundation. Appointment reminder calls and follow-up calls to patients that do not attend their appointments are effective in increasing breast cancer screening and diagnosis completion rates of (Goel A, et al, JHCPU 2008). We provided technical assistance to implement breast imaging care process improvements and compared pre and post process improvement appointment no-show rates and follow-up call rates at 14 sites. Procedures: Using Deming9s PDCA cycle for continuous improvement, we created care process improvement recommendations for 27 Chicago institutions with the patient base averaging more than 50% medically underserved patients (22 community, 3 academic and 2 public hospitals), 17 of the sites are safety net institutions. Recommendations were based on analysis across sites, literature, and input from institution staff. We provided 14 of the sites with technical assistance to implement “rapid cycle” care process improvements (RCI sites). Sites were selected for technical assistance based on their need and their commitment to improving breast imaging care processes. Thematic and statistical analyses were performed using simple frequencies, and McNemar9s test. Summary of Results: Detailed results are shown in the table below. Technical assistance was utilized by 14 of 17 safety net sites (RCI sites) to which it was offered. 10 sites adjusted their processes to conduct appointment reminder calls to patients. 12 sites implemented processes to conduct follow up phone calls with patients who missed an appointment, with a focus on answering patient questions and to schedule a new appointment. 6 sites adjusted their processes to conduct phone calls to inform patients of abnormal breast screening results (BIRADS 0, 4, 5) in addition to sending each patient a letter (MQSA requires at least a letter). Working with the 14 sites resulted in an improvement in no-show rates from 29% to 21%. Appointment reminder call: baseline 48% (13/27) all sites, 21% (3/14) RCI sites post-improvement 85% (23/27) all sites, 93% (13/14) RCI sites pvalue = 0.0044, Chi squared: 8.100 w 1 degree of freedom Call no-show patients : baseline 19% (5/27) all sites, 0% (0/14) RCI sites post-improvement 63% (17/27) all sites, 86% (12/14) RCI sites pvalue = 0.0015, Chi squared: 10.083 w 1 degree of freedom Call to follow up with patients who have abnormal results: baseline 56% (15/27) all sites, 50% (7/14) RCI sites post-improvement 78% (21/27) all sites, 93% (13/14) RCI sites pvalue = 0.0412, Chi squared: 4.167 w 1 degree of freedom No show rate for screening mammograms baseline 22% all sites, 29% RCI sites : post-improvement 18% all sites, 21% RCI sites Conclusions: Conducting phone calls to remind patients about appointments improves no-show rates at sites that care for the medically underserved. Further analysis may show a reduction in loss-to-follow-up for sites that implement follow up phone calls to patients who miss appointments and to patients who need additional diagnostic assessment. Findings from this study have generalizable application to health facilities beyond breast imaging sites. Given persistent and growing disparities in health outcomes for vulnerable populations and the limited resource availability; these findings suggest that implementation of basic patient tracking strategies have substantial benefit to improve patient outcomes and health care quality. Citation Format: Christine B. Weldon, Teena L. Francois, Julia R. Trosman, Betty Roggenkamp, Danielle M. Dupuy, Jimmie T. Knight, David A. Ansell, Anne Marie Murphy. Do patient follow-up improvements, at hospitals caring for medically underserved patients, impact no-show rates. [abstract]. In: Proceedings of the Sixth AACR Conference: The Science of Cancer Health Disparities; Dec 6–9, 2013; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2014;23(11 Suppl):Abstract nr A84. doi:10.1158/1538-7755.DISP13-A84

Illinois law opens door to kidney transplants for undocumented immigrants

Health affairs (Project Hope), 2015

Bias, black lives, and academic medicine

The New England journal of medicine, Jan 19, 2015

Federal anti-patient-dumping provisions: the first decade

Annals of emergency medicine, 1996

Colorectal cancer screening: trends in an urban primary care setting

American Journal of Gastroenterology, 2002